Bivariate Lifetime Model and Validation Procedure for Organic Light-Emitting Diode Displays

Abstract

Despite the advantages of organic light-emitting diode (OLED) displays over liquid crystal displays, OLED displays suffer from reliability concerns related to luminance degradation and color shift. In particular, existing testing schemes are unable to reliably estimate the lifetime of large OLED displays (i.e., displays of 55 inches or larger). The limited number of test samples and the immature technology result in great hurdles for timely product development. This study proposes a statistical approach to develop a lifetime model for OLED panels. The proposed approach incorporates manufacturing and operational uncertainties, and accurately estimates the lifetime of the OLED panels under normal usage conditions. The proposed statistical analysis approach consists of: (1) design of accelerated degradation tests (ADTs) for OLED panels, (2) establishment of a systematic scheme to build bivariate lifetime models for OLED panels, (3) development of two bivariate lifetime models for OLED panels, and (4) statistical model validation for the heat dissipation analysis model for OLED TV design. This four-step statistical approach will help enable accurate lifetime prediction for large OLED panels subjected to various uncertainties. Thereby, this approach will foster efficient and effective OLED TV design to meet desired lifespan requirements. Furthermore, two bivariate acceleration models are proposed in this research to estimate the lifetime of OLED panels under real-world usage conditions, subject to manufacturing and operational uncertainties. These bivariate acceleration models take into account two main factors—temperature and initial luminance intensity. The first bivariate acceleration model estimates the luminance degradation of the OLED panel

the second estimates the panels color shift. The lifespan predicted by the proposed lifetime model shows a good agreement with experimental results. Ensuring the color shift lifetime is a great hurdle for OLED product development. However, at present, there is no effective way to estimate the color shift lifetime at the early stages of product development while the product design is still changing. The research described here proposes a novel scheme for color shift lifetime analysis. The proposed method consists of: (1) a finite element model for OLED thermal analysis that incorporates the uncertainty of the measured surface temperature, (2) statistical model validation, including model calibration, to verify agreement between the predicted results and a set of experimental data (achieved through adjustment of a set of physical input variables and hypothesis tests for validity checking to measure the degree of mismatch between the predicted and observed results), and (3) a regression model that can predict the color shift lifetime using the surface temperature at the early stages of product development. It is expected that the regression model can substantially shorten the product development time by predicting the color shift lifetime through OLED thermal analysis.